In case of Chi Square: 16 cells (53.3%) have expected count less than 5. The minimum expected count is 1.58. So, it is not possible Dr Rabin Thapa. Thank you very much for your nice comments and suggestions Dr Rabin Thapa
I do not understand why you are reporting p values. These are descriptive statistics, not inferential tests of whether an independent variable is related to a dependent variable.
When you compare the Mean ± SD for two groups, you use the T-test, but here you have more than two groups, so you can use one-way ANOVA to compare among Mean ± SD for more than two groups. For more information you can check the following link: https://brownmath.com/stat/anova1.htm#Terminology
I'm sorry, now I noticed the table clearly after I zoom in... There is a comparison between two groups, the first comparison is between Mean ± SD by T-test and the second comparison between the proportions is in a chi-square... You can use the Med-calculator even it needs to compare every two values ( it takes time but simple to use and accurate)...Or you can use SPSS.
Thank you very much for your nice suggestion and comments Dr Manal Hadi Kanaan but how can i do it? What the real/actual name of Med-calculator? I am waiting for your suggestions Dr Manal Hadi Kanaan
If you want to compare between 3 groups with parameteric data then you can use ANOVA test.
If the data are non parmetric the compression between the three groups can be done by KrusKal_Walis test.
But for your table none of the above applies as the column 1 is the sum of the column 2 and 3.
Actually you have to analyse the between column 2 and 3. For which you can use chi-square test. Or since the values are given in percentage, you can also do z test of proportion.